Opened 7 years ago

#17891 new defect

rational powers of real interval field element (containing 0) give NaN

Reported by: dkrenn Owned by:
Priority: major Milestone: sage-6.6
Component: misc Keywords:
Cc: Merged in:
Authors: Reviewers:
Report Upstream: N/A Work issues:
Branch: Commit:
Dependencies: Stopgaps:

Status badges

Description

We have

sage: RIF(-1,1)^QQ(1)
[.. NaN ..]
sage: RIF(-1,1)^RIF(1)
[.. NaN ..]

whereas

sage: RIF(-1,1)^ZZ(1)
0.?

More generally, we it works for integers

sage: RIF(-1,1)^ZZ(123456)
0.?

as well as for the rational or RIF-element 2

sage: RIF(-1,1)^QQ(2)
1.?
sage: RIF(-1,1)^RIF(2)
1.?

but not for any other rational or RIF-element, e.g.

sage: RIF(-1,1)^QQ(3)                              
[.. NaN ..]
sage: RIF(-1,1)^QQ(4)
[.. NaN ..]

The reason seems to be

sage: log(RIF(-1,1))
[.. NaN ..]

The NaN-answer can be considered as wrong, since the true result is a number and can be calculated; it's a limitation of the algorithm.

Change History (0)

Note: See TracTickets for help on using tickets.